All references cited in this specification, and their references, are incorporated by reference herein where appropriate for teachings of additional or alternative details, features, and/or technical background.
Disclosed is an electronic oscillator system that provides accurate multiple phases of an oscillation. Oscillation, in electrical sense, can be defined as repetitive variation of voltage or current in time, which can then be used as signals to accomplish certain tasks, say in electronic devices such as mobile phones. Though prior art is abound with many different types of multiple-phase oscillators used as a source of timing signals for many electronic systems such as microprocessors, network processors, wireline/wireless transceivers and other data communication circuits, they generally suffer from unwanted irregularities in the amplitude and/or frequency of these signals. Even the slightest irregularities may become quite significant in very high-frequency high-accuracy type electronic circuitry. The presently disclosed traveling distributed wave oscillator can mitigate not only these types of unwanted discrepancies, but it can also sustain the oscillations indefinitely by a judicious use of an auxiliary oscillator as described later in the embodiments of a trigger-mode distributed wave oscillator.
Generally, there are two main types of electronic oscillators that produce repetitive electronic signals: the harmonic oscillator and the relaxation oscillator. The harmonic oscillator produces a sinusoidal output. The basic form of a harmonic oscillator is an electronic amplifier with the output attached to a narrow-band electronic filter, and the output of the filter attached to the input of the amplifier. When the power supply to the amplifier is first switched on, the amplifier's output consists only of noise. The noise travels around the loop, being filtered and re-amplified until it increasingly resembles the desired signal. The relaxation oscillator is often used to produce a non-sinusoidal output, such as a square wave or sawtooth. The oscillator contains a nonlinear component such as a transistor that periodically discharges the energy stored in a capacitor or inductor, causing abrupt changes in the output waveform. Square-wave relaxation oscillators can be used to provide the clock signal for sequential logic circuits such as timers and counters, while the sawtooth oscillators can be used in the time-based circuits that generate the horizontal deflection signals for cathode ray tubes in analogue oscilloscopes and television sets.
Most conventional electronic oscillator circuits use two reactive components, an inductor and a capacitor to create a resonant circuit, in an ideal case indefinitely transferring the energy from one to the other. However, in reality, the loss mechanisms associated with these reactive devices (can be modeled as resistance (R) and/or transconductance (G) elements) require active amplifying circuitry to compensate for these losses. The classical implementation for such an active compensation circuit is negative resistance circuit formed by cross-coupled active devices. A well-known MOSFET (Metal-oxide Semiconductor Field Effect Transistor) implementation of this configuration is shown in FIGS. 1a-1b and FIG. 2. The resultant oscillation frequency is given by the well-known relationship,
            f      osc        =          1              2        ⁢        π        ⁢                  LC                      ,where L is the inductance of inductor 30 and C is the capacitance of capacitor 40 of the so-called L-C tank oscillators 10 and 80 shown in FIGS. 1a-1b. The parasitic resistance R 20 associated with inductor 30 and the parasitic conductance G 50 associated with capacitor 40 results in losses in the tank-oscillators 10 and 80 of FIGS. 1a-1b, respectively that need to be compensated through cross-coupled inverting amplifiers 60 and 70 shown in the same FIGS. 1a-1b. In a CMOS (Complementary Metal-Oxide Semiconductor) technology, an inverting amplifier is implemented by connecting the drains and gates of a PMOS and an NMOS transistor together, resulting in a well-known oscillator circuit 80 shown in FIG. 1b, where 90 provides power supply potential (VDD). The devices 70 and 70′ of circuit 80 in FIG. 1b correspond to the inverting amplifier 70 of circuit 10 in FIG. 1a, whereas the devices 60 and 60′ of circuit 80 in FIG. 1b correspond to the inverting amplifier 60 of circuit 10 in FIG. 1a. 
Since the transmission lines are effectively distributed LC structures, distributed LC-oscillators can be constructed using transmission lines of which FIG. 2 is exemplary. A transmission line is, in general, parallel running conductors separated by a dielectric material. Micro-strip line (FIGS. 3a-b), coplanar wave guide (FIGS. 4a-b), coplanar strip line (FIGS. 5a-b), and differential coplanar wave guide (FIGS. 6a-b) are some of the most common transmission line structures. (Similar numerals refer to similar parts shown in FIG. 3-6. Thus 160, 170, 180 refer to respective signal lines, ground planes and dielectric layers separating the signal layer from the ground plane. Positive and negative signal lines are designated as 160+ and 160′, respectively. Similarly, character (a) references top-view, while character (b) references cross-sectional views of the respective transmission line structures in FIGS. 3-6). Although any of these structures can be used to construct an oscillator, the differentially symmetric ones are more favorable since the opposite phases of a signal are already available (coplanar strip line and differential coplanar wave guide).
Oscillator 100 in FIG. 2 shows an electrical model for a differential transmission line. In the same figure, 110 is inductance of Lodz/2 where Lo is inductance per unit length, 120 is resistance of Rodz where Ro is resistance per unit length, 130 is differential capacitance of Codz where Co is capacitance per unit length and 140 is differential conductance of Godz where Go is differential conductance per unit length for a differential transmission line stretching in z direction. The inductance per unit length and capacitance per unit length determine the phase velocity of the propagating wave. The phase velocity of a wave is given
  v  =      1    /                            L          O                ⁢                  C          O                    where Lo and Co are inductance per unit length and capacitance per unit length, respectively. Then, for a given total length of transmission line, the oscillation frequency can be calculated to be
            f      osc        =          1                                    L            tot                    ⁢                      C            tot                                ,where Ltot and Ctot are the total inductance and total capacitance along the transmission line. As described before, cross-coupled active amplifiers 150 are used to compensate for the conductor and substrate losses. Thanks to the distributed nature of these transmission line oscillators, multiple phases of an oscillation are available along the transmission line, whereas only two 180 deg opposite phases are available in case of a lumped L-C tank oscillators. Distributed Wave Oscillators, Rotary Traveling or Distributed Wave Oscillators, Standing Wave Oscillators are different classes of existing transmission line based oscillators all utilizing the distributed L-C nature of a transmission line structure.
FIG. 7 shows a simplified distributed oscillator of transmission line type 200 with characteristic impedance of Zo. The actual shape can be in any closing geometric form bringing point A to the vicinity of point B so that dashed AC coupled connection 210 can be obtained using a capacitor Cbp 220. The reflections resulting from the mismatch of the biasing resistor, Rmatch 230 to the line impedance, Z0, can be significant source of disturbance in the steady-state oscillation waveforms. This affect together with an additional non-ideality due to the bypass capacitor Cbp are the main drawbacks of this oscillator technique.
Another transmission line oscillator approach, Rotary Traveling Wave Oscillator technique shown in FIG. 8, avoids this disadvantage by direct cross-coupling 240 of the end points with an additional cost of odd symmetry introduced by this crossing of the transmission lines. The single-wire closed-loop structure of a Rotary Traveling Wave Oscillator limits the disturbances to one crossover which can still be significant at especially high-frequencies. Once enough gain is provided, there is no latch-up danger for the technique; since it utilizes a single-line DC-coupled closed-loop structure.
Standing Wave Oscillators (SWO) are another group of transmission line oscillators that would utilize transmission line structures. As is known by those skilled in the art, standing waves are formed by superimposing the forward and the backward distributed waves on the same transmission medium simultaneously. The two basic Standing Wave Oscillator topologies, quarter-wave λ/4 SWO 250 and half-wave λ/2 SWO 255 are shown in FIGS. 9a-b, respectively. A λ/2 SWO is basically combination of two λ/4 SWOs around a center symmetry point, with fundamental operating principle staying the same. In this type of oscillators, the differential transmission line structure is driven by cross-coupled amplifier 150 pair at one end, whereas the other end 260 is shorted. The waves created at the amplifier end 150 are reflected back at the short end 260 causing a reverse propagating wave along the transmission line. In the steady state, the forward and reverse waves coexist, creating standing wave along the line. This would imply amplitude variations in the oscillation phases along the line, gradually diminishing and eventually reaching zero at the short end 260′.
Circular Standing Wave Oscillator (CSWO) 270, shown in FIGS. 10a-10b, is still another standing wave type that would not require any reflection mechanism, but, rather a circular symmetry to create reverse propagating waves along the transmission line medium. As shown in FIG. 10a, the energy is injected into a closed-loop transmission line structure equally and travels symmetrically along the ring in clockwise 280cw and counter-clockwise 280ccw directions. These counter-traveling waves create standing waves with an amplitude profile as shown in FIG. 10b. It will be noted that where the wave components cancel each other a “quiet” node 290 is formed and a “loud” node 295, when the wave components reinforce each other. The energy is injected at two opposite points (A and B) with additional dashed connections 285 to force the main mode. Additionally, at least one of the quiet ports 290 has to be shorted to prevent any latch-up problems. This reduces this structure also to a single-line structure.
It will be known to those skilled in the art that conventional Standing Wave Oscillator structures have a critical drawback of amplitude variations which permits their usage to a limited set of applications. The oscillation phases corresponding to the quite ports would not even exist, compromising the main advantage of transmission line oscillators. What is needed, therefore, is an electronic oscillator circuitry that can provide invariant multiple phases of an oscillation in an uninterrupted manner.